Infinitival TO as a cohesion marker

HAL Id: halshs-00643791

https://halshs.archives-ouvertes.fr/halshs-00643791

Submitted on 14 Apr 2012

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Infinitival TO as a cohesion marker

Geneviève Girard-Gillet

To cite this version:

Geneviève Girard-Gillet. Infinitival TO as a cohesion marker. Recherches Anglaises et Nord Ameri- caines, Presses Universitaires de Strasbourg, 2011, pp.47-66. �halshs-00643791�

Infinitival to as a cohesion marker

Geneviève Girard-Gillet EA PRISMES

Université Sorbonne Nouvelle Paris 3 PRES USPC

Cohesion has something to do with a text, which is, according to Halliday and Hasan,

"a unit of language in use. It is not a grammatical unit, like a clause or a sentence; and it is not defined by its size" (Halliday and Hasan 1976: 1).

When analysing infinitival clauses, we do not generally consider that they play a role in the cohesive dimension of the text they appear in, but I think that this cohesive dimension might be relevant to our understanding of their function, as infinitival clauses are dependent clauses that derive their interpretation from the matrix clause or from the left co-text, and as such pertain to the general organisation of the text.

Halliday and Hasan insist that "cohesion occurs where the interpretation of some element in the discourse is dependent on that of another. The one presupposes the other, in the sense that it cannot be effectively decoded except by recourse to it. When this happens, a relation of cohesion is set up, and the two elements, the presupposing and the presupposed, are thereby at least potentially integrated into a text. (o.p : 4)

The fact that we can interpret : I'd like to, with to standing for something that was mentioned before, such as go to the theatre suggests that to has been substituted for a Verb Phrase and therefore plays some kind of role in the construal of the reference.

Even if Halliday and Hasan claim that "there are no structural relations between sentences", and therefore "ignore intra-sentence presupposition", (o.p.: 146) I think that to cannot be interpreted

- within a complex sentence, without taking into consideration the whole semantic organisation of the sentence, and

- within a text, without relating the missing elements, if any, to the left co-text.

In so doing, to seems to function as a discursive cohesion marker. After a brief survey of previous work, we'll explain what we mean by semantic cohesion in complex sentences, then focus on semantic cohesion with the left co-text.

I shall analyse cohesion from a semantic and a syntactic point of view, the two being often intermingle, but I won't have time to discuss gerundive/participial clauses. I already discussed some oppositions elsewhere, and plan to go into them in detail in the future.

1. Previous analyses

There is considerable disagreement among linguists as to the sense of the to infinitive.

Here is a brief overview of their hypotheses, opposing to infinitive and –ING. It shows a more and more abstract approach from Kiparsky and Kiparsky to cognitive linguistics.

The following table is adapted from Egan

Authors to infinitive -ing

Kiparsky and Kiparsky (1971) non-factive factive

Adamczewski (1978) phase 1 phase 2

Cotte (1982) devirtualisation

Quirk et al (1985) potentiality performance

Chuquet (1986) Choice of one value p or

p'

presuposition

Wierzbicka (1988) vague futurity vague simultaneity

Dixon (1991) potentiality activity extended in time

Duffley (1992) subsequent

potentiality/actualization

interiority

Smith &Escobedo (2001) conceptual distance conceptual overlap

Duffley (2006 ) possible movement

leading to the actualization

conceptualization of an event as an interiority

Deschamps (2006) visée préconstruction

Egan (2008) targeted alternative situation profiled as extended

I could also have mentioned Sweet (1903) and the opposition between to, specific, and ING generic; Jespersen and Wood, who adopted Sweet's conclusions; Hamawand with a contrast between to less immediate, and ING more immediate.

In spite of the varied conclusions reached, the authors are all searching for a semantic invariant, or core meaning, which could account for all the uses in context. I am not sure that this is an entirely satisfactory line of reasoning, since the numerous occurrences analysed thanks to the corpora available nowadays allow for more and more subtle descriptions, which is a real progress, but at the same time give rise to such abstract explanations that we can't but wonder what has been gained as regards the way non-finite clauses function. I tend to rather agree with Huddleston and Pullum, who say that "a single form does not always convey the same meaning and the same kind of meaning can be expressed by very different formal means." (2002:117)

I am not suggesting here that I have come up with a better explanation. I'm only trying to shed some light on some parameters, which, to my knowledge, have not been sufficiently considered.

2. Semantic cohesion within complex sentences.

Usage-based analyses, and in particular Thomas Egan (2008), show that only 70 to 75% of infinitives are compatible with a forward-looking interpretation. The 30% remaining cases concern complements to "believe-type" verbs, raising constructions, and the construction of aspectual verbs, among others¹.

2.1. The infinitive is the complement to a "want-type" verb.

Thomas Egan's comprehensive study mainly takes complement clauses into consideration. And it is true that infinitival complement clauses tend to denote events, situations, located after the moment expressed by the matrix verb. His approach takes into account the temporal interpretation of the infinitives, and the gerundives and shows that a great deal can already be accounted for, thanks to the temporal parameter.

Here are a few exemples of a forward-looking interpretation for to-infinitives :

1 See De Smet for the moments when the various constructions appeared.

(1) She closed her eyes, willing herself not to flinch. But she didn't draw back.

(P.D.James, The Private Patient, p 35)

(2)) She had expected him to be wearing a dark suit. Instead he was in very pale grey tweed. (PP, p12)

(3) If Candace now intended to leave, the sooner he knew the better. (PP, p 76) (4) 'Can't you see that I'm terrified that she might be planning to chuck me? (PP, p28) (5) The new Mrs Philpot would of course want to be mistress of her own home, and fill the house with children. (T.Chevalier, Remarkable Creatures, p 15)

(6) He had done what he had longed to do. (A.Sillitoe, The Open Door, p 358) (7) He knew exactly what she craved to hear. (OD, p 233)

What is worth noting here is that in all the examples the main verb already suggests itself a projection into the future. We then have a semantic and temporal cohesion which it is difficult to link to the presence of to. To does not seem to contribute in any way to this temporal interpretation. This can be proved by the future interpretation also derived from the past participle when the same matrix verb is used :

(8) I want my car fixed as soon as possible.

It is the meaning of want which forces this temporal construal, and –EN does not bring its own temporal meaning to the whole structure, apart from the passive interpretation concerning my car fixed.

Here is another example :

(9) There are some tests I would like done, including blood tests and an EGG, but I would prefer those to be carried out at St Angela's. (PP, p 14)

The use of ING after want is another proof that to is not per se the marker of a posterior interpretation. In (9) we have two different complementations : V-ing and to+ V, but each clause is interpreted as referring to a future process.

In example (10) want is first followed by ING, then by to. In both cases the father worries about what the future holds in store for his son.

(10) He doesn't want Alex' standing out from the crowd. He knows that soon the boy's life will become difficult and he hopes conformity might be his saviour. He wants him to be part of this everybody. (Z.Smith, The Autograph Man, p 6)

I'm not saying that to + V and V-ing are interchangeable, but the temporal interpretation is the same.

To is compatible with the future interpretation already present in the governing verb, and as such can't be said to have a meaning of its own. It would, if it forced a different reading. It seems better then to say that it is semantically "transparent", and that it lets the temporal interpretation denoted by the main verb expand onto the infinitival clause.

The verbs are activity verbs, requiring, if the activity alluded to is actually performed, the Agentivity of the subject. This agentivity confirms the prospective reading, as the process can't be simultaneous with the moment encoded by the matrix verb. If it were, no desire that it should occur could be logically expressed.

This notion of "transparency" is indirectly suggested by Smith and Escobedo, who after saying that "to often evokes the idea of futurity", suggest that "to is partially motivated with the verbs in this set ("want" verbs) because its future sense overlaps semantically with the forward-looking meaning of the matrix verbs." (2001:554)

2.2. Adjectives governing infinitives

Let us see now what interpretations we get when the governing element is an adjective.

(11) He said, anxious not to sound intrusively curious but needing to know, 'will you want to stay here for some time?' (PP, p 79)

(12) I was ready to tumble into a ditch and make a run for it. (OD, p 144)

(13) Now that she was at last free to leave, Mrs Skeffington became gracious and bestowed vague thanks and insincere smiles indiscriminately on everyone present. (PP, p155) (14) At the door he was impatient to get away. 'Remember me to Aunt Lydia.' (OD, p 55)

In these four occurrences the semantic content of the adjective denotes a psychological state (anxious, ready, impatient) or a physical one (free) concerning a future situation which is expressed by the infinitive.

On the contraty, in the next examples, the psychological state presupposes a past or a present situation, which has triggered the feeling of relief, of shame, of pleasure :

(15) He was relieved to hear her voice. (OD, p 168) (16) 'When you've finished I'll show you your bed.'

He hoped she would go, and was ashamed to be seen taking his tunic off, which he was, in front of a stranger. (OD, p 81)

(17) I don't think patients who come here for privacy would be delighted to find themselves in the company of a notorious journalist. (PP, p 81)

(18) No doubt the people here will be glad to see the body leave. And then we can have a word before I go.' (PP, p141)

The infinitive could be deleted, but the interpretation of the feeling would nonetheless suggest that there was, there is a particular situation giving rise to the feeling in question. We have a kind of cause-consequence interpretation :

[To hear her voice] caused him to feel relieved

[To be seen taking off his tunic] caused him to feel ashamed

This interpretation does not tie up with the adjectives we have in (11)-(14) :

*[to sound curious] caused him to be anxious

*[to tumble into the ditch] caused him to be ready]

In all the occurrences, the subject of the infinitive is co-referential with the subject of the governing element, which reinforces the cohesion. In cases of non co-reference, a finite clause is compulsory.²

2 I shall not treat raising constructions here, but they are temporally constrained in the same way. The infinitive in John seems to be depressed can only mean that he looks depressed now, at Speech time. It can't mean that the

2.3. The infinitive is the complement to a "believe- type" verb.

To seems to be transparent, and having no time marker of its own allows for the infinitive to adopt the temporal reading denoted by the semantic content of the governing verb or governing adjective.

But verbs of judgment do not project into a future period, but express the subject's viewpoint at a certain time. It stands to reason then that the infinitive will not highlight a posterior situation, but on the contrary focus on the situation at the moment assessed through the main verb. Such verbs function as epistemic modals, to a certain extent.

(19) Clara had made it plain that she considered him to be too old, too absorbed in his job and his poetry to make a serious commitment to any woman. (PP, p112)

(20) Maybe she thought his remarks were the narrow outlet of a full reservoir, and that he wasn't as empty as she supposed –or a he knew himself to be. (OD, p264)

(21) Now tell me, as a teacher, as an English specialist, as a man of letters, what exactly you take "quite satisfied" to mean. (A.S.Byatt, The Virgin in the Garden, p 37)

(22) She assumed him to be a worse case than he was, but there was no word for it that he liked. Tuberculosis assumed you were ill, while consumption meant you were dead. (OD, p 263)

The lack of specific interpretation for the infinitive, and its interpretation as past or present, depending on the tense in the matrix verb is another case of cohesion.

The infinitive "time" is simultaneous with the moment when the judgment is passed : present, past, or future. Since judgment verbs do not have, within their semantic features, a temporal orientation, unlike "want-type" verbs, only the tense marker they carry play a temporal role, and the infinitive interpretation can only be a temporal interpretation by default. This interpretation matches that of the main verb.

Unlike the lexical verbs in 2.1. the verbs are stative or the copula.

So I can't agree with Duffley, who says : "The potential meaning of to before the infinitive is more abstract than that found in the spatial use of the preposition, and can be stated as follows: the possibility of a movement from a point in time conceived as a before-position to another point in time which marks the end-point of the movement and which represents an after-position with respect tothe first." (1992:16)

With judgment verbs the semantic domain of the infinitives is restrained to states or situations depicted by stative verbs (mean, for instance) or BE. The lack of dynamism in the semantic content of the clause might explain why we only get the simultaneous reading. It is worth noting though that other temporal links are possible, but in that case finite clauses are the only solution : everybody believes that he will change his mind / she believed that he had criticized her.

2.4. Role of the tense marker

The presence of a past tense on the matrix verb poses an interesting question, which I shan't have time to address in detail unfortunately.

situation will take place in the future, as *John seems to be depressed tomorrow in unfelicitous. We have to resort instead to a finite clause : it seems that he will be depressed tomorrow.

Let us only take an invented example, with no context at all, for convenience' sake : I expected Mary to arrive late.

The sentence can be analysed from two angles :

(a) I remember now the moment when I was waiting for Mary to arrive and at that time I thought that she would be late and that there was no point in baking a soufflé. I might have said to myself : she will be late, or I expect her to be late. In this structure the infinitival clause expresses the projection into a future moment of [her-to be late], since the temporal interpretation is derived from the meaning of expect.

(b) but a question immediately arises : why am I uttering such a sentence now ? Either because I want to relive the past situation, hence the (a) interpretation, or more obviously because I know now that Mary did not arrive late. What does this mean as regards the English structure, and the interpretation of the infinitive ? In case (a) the predicative relation [Mary-be late] is posited as non validated yet, and in case (b) the predicative relation [Mary-be late] is understood as non validated in the past. So the same syntactic structure, the infinitival clause, encodes two interpretations depending on the governing verb, plus the tense on this verb³. It is hardly possible then to suggest that the infinitive has only one meaning, since we need to analyse the semantics of the main verb and the tense it carries in order to reach a satisfactory reading of the whole sentence.

Another interesting verb is the verb forget. Why and when do we say : I forgot to post the letter ?

3. Semantic cohesion with the left co-text.

3.1. The infinitive is the subject of the matrix verb.

The infinitival clause appearing as the subject plays a completely different role : it enlarges, develops the topic of the left co-text, either with a specific reading, or a more generic one.

This point has already been pointed out and analysed by P.Miller (1999 : 105) who said that the preferred structure is the non-extraposition of the infinitive when its content can be derived or inferred from the left co-text⁴ :

"Pour apparaître en position sujet la complétive doit avoir été évoquée dans le discours précédent ou être directement inférable de celui-ci."

Here are a few occurrences :

(23) None of the patients in the waiting room except herself gave any clue to the particular expertise they required. To become a patient was to relinquish a part of oneself, to be received into a system which, however benign, subtly robbed one of initiative, almost of will. (PP, p 11)

The focus in the left co-text is on a patient waiting to be examined by her plastic surgeon. From the description of the waiting room and the other patients, we move on to a general assessment of what being a patient involves. This judgment to relinquish a part of

3The sentence can be followed by : and she did, but this is the marked interpretation.

4The extraposition is not always possible when the matrix verb is transitive, and excluded when be is identificational.

oneself can only be posterior to the mental representation construed first by the enunciator to be a patient.

To invert the two infinitives would not make any sense in the text, since it would mean that what was discussed before was the abandonment of oneself.

(24) She would come to the Manor as a patient and, for her, this implied vulnerability and a certain physical and emotional subservience. To arrive unbriefed was to put oneself at a disadvantage from the start. (PP, p 22)

The woman who will be operated on is advised to stay at least one day at the clinic (the Manor) a week or so before the operation to get used to the environment and the staff. She feels vulnerable, and thus agreed to go, not to arrive unbriefed on the day the operation is planned.

(25) She contemplated the coming operation without fear while recognizing that she had some rational cause for anxiety; to be given a general aenesthetic was never without risk.

(PP, p 31)

We have the same semantic link between the notion of fear and the idea that a general aenesthetic can be dangerous. The anxiety alluded to is the anxiety prompted by the risk always present in an operation.

(26) For over thirty years she had nursed her contempt, her resentment and, yes, her hatred. It was herself that she had harmed and to recognise this might be the beginning of healing. (PP, p 36)

The sentence it was herself she had harmed suggests that she is now aware of the pain she had inflicted on herself; this is summed up, as it were, by to recognize this.

(27) He had to make a decision –and make it now- whether to remain at the Manor and work with one of the best surgeons, or to go with a surgical team to Africa. […] But to leave him now would still be a betrayal of a man he revered. Not to leave him would be a betrayal of himself and of Candice, the sister who, loving him, knew that he had to break free. (PP, p 70)

The man, Marcus, who is contemplating his future career is the assistant of a famous Master of surgery, whom he feels ashamed of abandonning, but he knows that his deepest desire is to help people in Africa. The two subject infinitival clauses present the two options.

Such sentences are actually very frequent, at least in fiction. All the examples here are taken from The Private Patient, by P.D.James, her latest novel.

(28) Don't we all at some time or another make a decision which we know is absolutely right ? And even if it fails, to resist it will be a greater failure. (PP, p 79)

(29) She lay unmoving as objects solidified round her. How many hours, she wondered, had the operation taken? One hour, two, three ? She pondered the difference between this temporary death and sleep. To wake after sleeping, even a deep sleep, was always to be aware that time had passed. (PP, p 94)

The infinitive elaborates on the question the woman asked herself : how long had she slept, since she had no idea how long the aenesthetic had worked.

(30) At half past ten on that Saturday morning, Commander Adam Dalgliesh and Emma Lavenham had an appointment to meet her father. To meet a future father-in-law for the first time, especially with the purpose of informing him that one is shortly to marry his daughter, is seldom an enterprise undertaken without some misgivings. (PP, p105)

(31) 'Not bad news, I hope. Your parents are well ?'

'Both have died, sir.' That was an official call. I'm afraid that I shall have to leave urgently.' Dalgliesh feared that the Professor might comment that to lose one parent may be regarded as a misfortune, but to lose both looked very like carelessness, but it was apparent that there were some remarks that even his future father-in-law baulked at. (PP, p111)

In all the sentences, the infinitive enlarges the theme and has a generic interpretation leading to a jugdment. The jugdment presupposes a mental representation of the situation and then necessarily comes after this representation. The infinitive then encodes a mental anteriority, and consequently has a completely different interpretation than those described in 1.1 or 1.3.

when the infinitive was a complement to the main verb. Syntax then plays a fundamental role when it comes to interpreting infinitives, and this syntax-semantics linkage calls for more analyses.

In some cases though, the situation described by the infinitive has actually taken place or is the current situation. We'll see some examples in the next part.

3.2. The infinitive encodes an already actualized situation

(32) Her childhood had been defined for her by the words of her grandmother to a neighbour, 'If her ma had to have an illegitimate child, at least she could have stayed alive to look after it, not dumped it on me!' What remained with Kate, and always would, was the knowledge that never to have known either parent was to live life with some essential part of you missing, a hole in the psyche which could never be filled. (PP, 115)

Kate is an orphan, and what is expressed by the infinitive is her own experience.

(33) 'You have some scars on your left lung'

He was meant to understand something, but if that was a flicker of irritation on the medical officer's face, he needed more clues. To be considered dim pushed him even more into thoughtlessness.

'Scars, sir? ' (OD, p 8)

The man has just be diagnosed for TB, but doesn't really understand what the M.O. meant.

(34) He would apply for a passport, a step up in the world, a mark of identity –and an inducement to travel when to be in England felt like imprisonment. (OD, p 169)

(35) Such light provided a thread between the time when he had been too busy to think, and the time, which was now, when to think about what had struck him was the only way of staying alive. (OD, p 60)

Duffley (1992: 116) also mentioned cases denoting a realized event, especially in exclamations :

You unmanly fellow !to treat a woman so who took you off the street ! but does not seem to see the contradiction with his general hypothesis.

I can't agree either with Stowell's hypothesis : "its status as being neither present nor past has the effect of specifying that the time frame of the infinival clause is unrealized with respect to the tense of the matrix in which it appears. In other words, the tense of a to-infinitive is that of possible future". (1982 : 562)

Here are a few occurrences when the infinitive does not depend on a governing element, but functions as a kind of independent clause.

(36) 'And now you're dripping all over my clean floor. I was about to phone the guardroom and report you missing'

He wasn't a kid to be slammed in this way. If she said much more, he'd tell her where to go.

(OD, p 136)

The infinitive refers to what the character experienced. There is no syntactic linkage, but the reader construes a logical link where the "slamming" must be anterior to the judgment.

(37) The kitchen was cold, with the fire in the range feeble. […] Poor Mary, I thought. To be on the cold beach all day and come back to this. (RC, p 95)

A possible paraphrase is : to be on the cold beach and come back to this must be a terrible experience. And the reader knows that Mary spends all her days on the beach, collecting fossils.

(38) She didn't realize that her words slurred. He suspected that his did. Listening, he decided that they didn't, though if they did how would he detect it ? Being drunk, they would get to know each other. But he was surely drunk to think so. (OD, p 172)

The infinitive to think so refers to the thoughts expressed in the previous sentences, and these thoughts lead the male character to hypothesise that he must be drunk.

(39) That will teach you to buy a Japanese car ! (reference lost)

The sentence was pronounced by someone overtaking a Japanese car which was stuck on the hard shoulder waiting to be hauled out to a nearby garage. The derogatory remark implies that the man who bought the Japanese car had not made the best choice.

(40) That will teach me to be rational ! (A.Christie, The Mirror cracked from side to side)

Miss Marple suddenly understands that her rational remark was not relevant in the situation, and feels stupid⁵.

5 All these occurrences call for further research, as they haven't been studied yet in detail.

3.3. Cases with ellipses of the VP.

In my introduction I mentioned that to can replace a whole Verb Phrase, and as such functions as a modal or auxiliary. Such a role is possible since to is a free morpheme. An important distinction is to be made here with ING, which, as a bound morpheme, can't play this part. ⁶

(41) 'I can't stay here indefinitely, you said that yourself.'

'I'm not asking you to. I'm just asking you to consider carefully which way you want your future career to go.' (PP, p 74)

(42) 'If your sister has personal reasons for her antagonism to Miss Gradwyn she need have nothing to do with her. She's not required to, the patients are not her concern.' (PP, p 75)

(43) Li-Jin is saying, You are quite sure that you want to do it (see Big Daddy versus Giant Haystack in a wrestling match) ?

Alex-Li is saying, Da-aaad ! Alex, I'm asking you a question.

'I know. I said yes, didn't I? Well, then. Yes. I s'pose.'

'But are you sure you want to? Or could it be more that your mother wants it? (Z.Smith, The Autograph Man, p 6-7)

(44) She said, 'At present it's too dark to see anything but you'll be able to in the morning (PP, 40)

(45) 'But you weren't left alone at any time, were you? I arranged that you shouldn't be.'

'It was as bad as being alone. Just the one person. Sister, who didn't stay for long, wouldn't talk about what happened. I suppose she was told not to.' (PP, 149)

(46) we say very little, for we do not need to. We are silent together, each in her own world, knowing the other is just at the back. (RC, p 343)

(47) Whenever I brought up topics like the possible age of the earth , she twisted her hands in her lap and changed the subject as soon as it was polite to. (RC p 293)

(48) 'And you didn't hear the lift?'

'We didn't hear it. Nor would I expect to'. (PP, p172)

The fact that it is the left co-text that we retrieve thanks to the presence of to leads me to posit that to can't have a specific meaning, since this meaning would interfere with the interpretation, if it had one. The item that does play a role is each time the governing element.

Conclusion

An infinitival clause is an economical construction, whose interpretation relies greatly on the linguistic environment :

6 This has other far-reaching consequences.

-Its subject is not always expressed, depending on the semantic class the verb belongs to, but it has to be retrieved, or is understood as generic.

- it has no tense marker, but must nevertheless receive a temporal interpretation, which, we have seen, is determined by the semantic function it has in the whole sentence : when it is a complement, we construe a future interpretation with "want" verbs, but a simultaneous

one with "believe" verbs. When it is a subject, the interpretation is one of anteriority (real, or presupposed). It can be interpreted as referring to an actualized situation in the real world, or to a purely mental representation in the speaker's mind.

All these necessary data, which are not encoded in the infinitive, but which are to be taken into consideration, and which are dependent on the co-text (within the sentence, or in the left co-text) cast some doubts on the possibility of finding a core meaning that would be present in all the occurrences.

But these deducible data seem to suggest that the various interpretations can only be construed if some kind of cohesion is set up in the text.

Most of the examples are taken from : OD : The Open Door, by A. Sillitoe (1989) PP : The Private Patient, by P.D.James (2008) RC: Remarkable Creatures, by T.Chevalier (2009)

References

ADAMCZEWSKI H. (1978). Be+ing dans la grammaire de l'anglais contemporain, Champion, Paris.

BOULONNAIS D. (2004). « TO et les infinitives : l’hypothèse de la transcendance prépositionnelle », in La contradiction en anglais, CIEREC, Travaux 116, 55-90.

CHUQUET J. (1986) : To et l'infinitif anglais, Ophrys, Gap.

COTTE, P. (1982) : "To, opérateur de dévirtualisation", Modèles linguistiques T.IV 2, Presses Universitaires de Lille.

DE SMET H. (2008). Diffusional change in the English system of complementation; gerunds, participles and for … to infinitives? Ph.D dissertation, Universiteit Leuven

DESCHAMPS A. (2006). « Pour un traitement formel énonciatif des compléments de verbe à forme finie », in C.Delmas (ed), Complétude, cognition, construction linguistique, PSN, Paris.

DUFFLEY P. J. (1992). The English Infinitives, Longman, London.

DUFFLEY P. J. (2006). The English Gerund-Participle. Peter Lang.

DUFFLEY P. J. (2007). « A natural-language semantics approach to infinitival and gerund- participial complementation in English » in Anglophonia, Presses Universitaires du Mirail, 55-67.

EGAN T. (2006). « The Classification of Non-Finite Complement Constructions in English », in D.Lebaud, C.Paulin, K.Ploog (eds), Constructions verbales et production du sens, Presses Universitaires de Franche-Comté.

EGAN T. (2008). Non-finite Complementation. A usage-based study of infinitive and –ing clauses in English. Amsterdam/New York.

FREED Alice F. (1979). The semantic of English Aspectual Complementation. D.Reidel, Dordrecht.

GIRARD G. (1993). « Cease+toV/stop V-ing et la notion de sujet identique », in Sigma, n° 16 GIRARD-GILLET G (à paraître) « Aspectual Verbs : a study of « cease » and « continue », Actes du colloque Evénement/Sous-événement, PSN, Paris.

HALLIDAY M.A.K. & R. HASAN (1976). Cohesion in English, Longman, London.

HAMAWAND Z. 2002. Atemporal Complement Clauses in English : A Cognitive Grammar Analysis, Lincoln Europa, München,

HUDDLESTON R. & Pullum G.K. (2002). The Cambridge Grammar of the English Language, CUP, Cambridge.

JESPERSEN O. (1940). A Modern English Grammar, Part V, Allen and Unwin, London.

KHALIFA J.C. (2004). Syntaxe de l'anglais, Ophrys, Gap.

KIPARSKY P & C. KIPARSKI (1971). "Facts" in D.S.Steinberg & L.A.Jacobvits (eds), Semantics: An Interdisciplinary Reader in Philosophy, Linguistics and Philosophy, CUP, Cambridge.

LANGACKER R.W. (1987). Foundations of Cognitive Grammar, vol1. Standford University Press, Standford.

LARREYA P. (1987). "Peut-on porter un regard simple sur les formes impersonnelles du verbe anglais : (TO) + V, V-ING, V-EN ?" , Sigma n°11, CELA, Aix en Provence.

MELIS G. (2007). "Hypallage et incidence de l'adjectif épistémique dans les structures complexes en anglais contemporain", in Anglophonia, 22. Presses universitaires du Mirail.

MILLER P. (1999). "L'extraposition des complétives sujet et objet", in D.Roulland (ed), Actes du colloque de l'ALAES, Rennes, Université Rennes 2.

QUIRK R &al. (1985). A Comprehensive Grammar of the English Language, Longman, London.

SMITH M.B. & J. ESCOBEDO . (2001). "The Semantics of to-Infinitival vs.-ing Complement Constructions in English, in Andronis M &al (eds), Proceedings from the Main Session of the Thirty-Seventhh Meeting of the Chicago Linguistic Society, vol 37, n°1, Chicago Linguistics Society, Chicago.

SWEET H. (1903). A New English Grammar, Vol 2, Clarendon, Oxford.

TALMY L. (2000). Towards a cognitive semantics II : typology and process in concept structuring. Cambridge, MA : MIT Press.

WIERZBICKA A. (1988). The Semantics of Grammar, John Benjamins Publishing Company, Amsterdam.

WOOD F.T. (1962). Current English Usage, Macmillan.